Greetings!
Currently, I am conducting a meta-analysis on the association between BMI and prognosis. I discovered that the studies used various cut-offs. However, only two studies are eligible for dose-response meta-analysis (DRMA) as only these studies reported >1 cut-off values. Regarding this study:
1. Is it possible to perform a dose-response meta-analysis when only 2 studies are included?
2. Is there any statistical software that is able to perform dose-response meta-analysis other than STATA and R?
3. Regarding the preliminary analysis, we would like to perform a two-class analysis by comparing studies reporting 1 cut-off values (i.e. >=30 vs =25 vs
Define dose-response meta-analysis. I've have never heard of such. Perhaps you mean doing meta-regressions. Do you mean that the dose varies among studies or within each study? Regardless, with two studies, you can't do much in the way of doing a meta-analysis.
1. This would be similar to trying to establish linearity with only two data points. You are not able to conclude anything about a dose-response relationship, with two studies with different doses.
However, if you mean that the two studies have investigated a dose-response relationship, and you want to pool these data, then you are one step closer to say something about dose-response, but two studies is very little, and I would not do a meta-analysis.
2. My suggestion would be SAS, apart from STATA and R.
3. You could do both an overall analysis with high vs low BMI, and a stratified analysis with the two different cuttoffs, and present it in a forrest plot.
Of course it should be discussed as a limitation, if you conclude on the analysis pooling results from different cut-off values.
I hope that this was helpful.
With only 2 data points, it is not a very good option to do a dose-response MA.
if you have studies with data for several cutoff values, then I think it would be OK to do a DRMA. If you have other studies that have assessed only 1 cut-off and if you know that cut-off value, you can include them to if you can extract the relevant data you need.
In terms of software, if you are familiar with coding, then R, STATA, SAS, ...
If you aren't familiar with coding I would suggest openmetaanlyst (free and easy).
Openmetaanalyst allows you add create continuous and discrete co-variables (if you want to do a meta-regression based on the cutoffs, you should choose the continous and use the cutoffs reported; if you want to group them into 2 or more classes, choose the discrete co-variable). If you want to do 30 and 25 you would need 2 discrete co-variables, one for each comparisson.
Hope this helps
Good luck for your study.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software